The mothers of some autistic children may have made
antibodies against their fetuses' brain
tissue during pregnancy that crossed the placenta and
caused changes that led to autism, suggests
research led by
Hopkins Children's Center investigators and published
in the February issue of
the Journal of Neuroimmunology.
The causes of autism, a disorder manifesting itself
with a range of brain problems and marked
by impaired social interactions, communication disorders
and repetitive behaviors, remain unknown for
an estimated 90 percent of children diagnosed with it.
Genetic, metabolic and environmental factors
have been implicated in various studies of autism, a
disorder affecting one in 150 children in the
United States, according to estimates by the Centers for
Disease Control and Prevention.
"Now our research suggests that the mother's immune
system may be yet another factor or a
trigger in those already predisposed," said lead
investigator Harvey Singer, director of Pediatric
Neurology at Johns Hopkins Children's Center.
Researchers caution that the findings needn't be cause
for alarm but should be viewed instead
as a step forward in untangling the complex nature of
Mostly anecdotal past evidence of immune system
involvement has emerged from unusual
antibody levels in some autistic children and from
postmortem brain tissue studies showing immune
abnormalities in areas of the brain. Antibodies are
proteins the body makes in response to viruses and
bacteria, or sometimes mistakenly against its own tissues.
Yet the majority of children with autism
have no clinical evidence of autoimmune diseases, a fact
that prompted researchers to wonder
whether the antibodies transferred from mother to child
during pregnancy could interfere with the
fetal brain directly.
To test their hypothesis, the researchers used a
technique called immunoblotting (or Western
blot technology) in which antibodies derived from blood
samples are exposed to adult and fetal brain
tissue to check whether the antibodies recognize and react
against specific brain proteins.
Comparing the antibody-brain interaction in samples
obtained from 100 mothers of autistic
children and 100 mothers of children without autism,
researchers found either stronger reactivity or
more areas of reactivity between antibodies and brain
proteins in about 40 percent of the samples
obtained from the mothers of autistic children. Further,
the presence of maternal antibodies was
associated with so-called developmental regression in
children, increasingly immature behaviors that
are a hallmark of autism.
While the findings suggest an association between
autism and the presence of fetal brain
antibodies, the investigators say further studies are
needed to confirm that particular antibodies do
indeed cross the placenta and cause damage to the fetal
"The mere fact that a pregnant woman has antibodies
against the fetal brain doesn't mean she
will have an autistic child," Singer said. "Autism is a
complex condition and one that is likely caused by
the interplay of immune, genetic and environmental
Researchers are also studying the effect of maternal
antibodies in pregnant mice. Preliminary
results show that the offspring of mice injected with brain
antibodies exhibit developmental and
social behaviors consistent with autism.
The senior author on the study is Andrew W. Zimmerman,
of the Center for Autism and Related
Disorders at the
Kennedy Krieger Institute. Co-authors are Christina
Morris and Colin Gause, both of
the Johns Hopkins School of Medicine; Pam Gillin, of the
Kennedy Krieger Institute; and Stephen
Crawford, of the Johns Hopkins Bloomberg School of Public
Health. The study was funded by the
Alliance for Autism Research.